Tunnelling

Management by Design

Alan Muir Wood

• Publisher: CRC Press
• Year: 2000
• ISBN:
  9781135809492
• Edition: 1
• Topic: Architecture
• Subtopic: Architecture Methods & Materials

Tunnelling has become a fragmented process, excessively influenced by lawyers'notions of confrontational contractual bases. This prevents the pooling of skills, essential to the achievement of the promoters' objectives. Tunnelling: Management by Design seeks the reversal of this trend. After a brief historical treatment of selected developments, th

Chapter 1
Background to modern tunnelling

If we can think of one bit of time which cannot be sub-divided into even the smallest instantaneous moments, that alone is what we can call the ‘present’. And this time flies so quickly from the future to the past that it is an interval with no duration.

St Augustine, Confessions XI.xv (20), tr. H.Chadwick.

1.1 Introduction

The reader will find here no attempt to provide a comprehensive history of tunnelling. The objectives are more modest and more targeted, namely, to set the scene from the earliest known tunnelling, to review developments which have contributed to present-day understanding of the criteria for success, with references to exemplary projects.

On the negative side, the objective is to attempt to explain historical factors contributing to the current lack of general appreciation for these same criteria for success. In so doing, it is necessary to understand how circumstances have changed with time and with the increased complexity of the tunnelling operation. The many contributory factors include the increasingly specific technical requirements to satisfy the objectives of an underground project, developments in the techniques and the means of tunnelling, also in the associated improved capability to explore the ground, to measure and model its characteristics.

This account is in consequence deliberately selective, using particular examples, from the several periods of development of tunnelling, as landmarks, generally to emphasise the varying degrees of intuition, craft-skill and technology which have characterised the several periods into which this chapter is divided. There is no clearly defined boundary between tunnelling history and current developments. The most important features of current practice rest upon concepts which have been developing over many years. The sub-division between past and present is therefore one of convenience in telling a reasonably continuous tale.

A broad survey of tunnelling rapidly reveals how tunnelling has progressively been challenged to penetrate ground of increasing degrees of inherent difficulty. With time, in consequence, the relationship between the main parties concerned with each tunnel has assumed greater importance. A contrary more recent development, during a period in which inappropriate contract relationships have been increasingly imposed, adds to a mismatch between what these relationships should be and what in fact they have been. A pervading objective of this book is to help to guide towards good practice for the future, in the management aspects of engineering as much as the technology, the two being indivisible in good practice. A historical perspective provides a guide, with the view that knowledge of the point of departure is necessary in order to chart the course to our destination. For general accounts of the history of tunnelling, the reader may turn to Sandström (1963), Széchy (1970) and Harding (1981) although it will be found that more recent information causes the account below to make some modification to these texts. Background reading into earlier times may start from Singer et al. (1954).

1.2 Tunnelling in antiquity

Since tunnelling, especially tunnelling for mining minerals and for water supply, precedes recorded history, we can only conjecture as to the thought processes of those in the earliest times who discovered that tunnelling provided the solution to perceived requirements. We need to respect the intelligence of these early tunnellers of 3000 years ago and more for the magnitude of their achievements, whose art depended entirely on trial-and-error (heuristics) to learn what could be done and how best to achieve results, with neither the benefit of underlying technology nor with access to specialised tools. All ancient civilisations have left behind examples of tunnelling at varying scales.

The earliest tunnels would have been modelled on natural caves, selecting rock types capable of penetration with crude tools, and having innate strength and tight joint structure conducive to a natural stable form without need for artificial support. Much of the tunnelling for mining was initially of this nature following lodes with near-surface exposures. Early tunnels reflect the change of social priorities with time. Whereas at the present day a tunnel has one or more special purposes as conveyor of people or substance, storage or protected housing, many of the most spectacular tunnels were for ceremonial and religious purposes. It is remarkable that the first known subaqueous tunnel was built in about 2000 BC to connect the royal palace of Queen Semiramis to the Temple of Jove (or his equivalent) beneath the River Euphrates. The tunnel was about 1 km long, had a section of 3.6 m×4.5 m, and was lined in brickwork with a vaulted arch set in bituminous mortar. The river was diverted to permit cut-and-cover construction. While this may not constitute tunnelling in the purest sense, nevertheless the sophistication of the techniques point to a preexisting skill and assurance in undertaking such works.

It also deserves record that, for city drainage systems during the Akkadian supremacy (BC 2800–2200) of Mesopotamia, vaulted sewers were built in baked brick, with inspection chambers and rodding eyes, lined in bitumen. To wander a little further from our scene, there were bath-rooms paved in bitumen-faced brick, also closets with raised pedestals with the occasional refinement of a shaped bitumen seat (Singer et al. 1954). Rings of brickwork in vaults for wide spans were inclined to avoid the need for centring.

Tombs and temples provide other examples of tunnelling. The Royal Tombs at Thebes in Egypt and at Ur in Mesopotamia date from around 2500 BC; Abu Simbel in Egypt dating from around 1250 BC is another such example. These tunnels were in limestone, requiring high standards of craftsmanship in their form but presenting no problems of stability. Another later example, mentioned by Sandström (1963) among many examples in India, concerns the caves of Ellora, near Bombay, excavated between 200 and 600 AD, tunnels aggregating to a length of more than 9 km, cut by chisel in fine-grained igneous rock.

A celebrated, but by no means unique, water supply tunnel of the ancient world is the Siloam tunnel (also known as Hezekiah’s tunnel), mentioned at 2 Kings 20:20, at 2 Chronicles 32:30 and elsewhere in the Old and New Testament of the Bible. The oldest part of the city of Jerusalem, dating from 2000 BC or earlier had a source of water from the Gihon spring on the Ophel ridge to the east of the city. The name Gihon in Hebrew (and its Arabic form, Umm al-Darah) relates to its gushing siphonic nature, with high flows for a duration of 30 minutes or more at intervals of 4–10 h (depending on the season and the source of information). As early as about 1800 BC, a short length of tunnel connected the spring to a well shaft nearly 30 m deep within the city wall. In around 950 BC Solomon connected the spring by an open channel outside the city wall to the pool of Siloam within the wall. Under threat of siege by the Assyrians under Sennacherib, King Hezekiah secured the water supply by resiting the pool of Siloam and feeding it through a tunnel 533 m long, following a sinuous route about 60% longer than the direct length from spring to outlet portal. The tunnel is roughly 1 m wide and 2 m high, increasing to a considerably greater height near the portal, presumably as a result of correcting an error in level. The reasons for the devious course of the tunnel have long been the cause of speculation. Most probably the explanation includes geological influences (natural fissures in the rock) combined with the avoidance of areas of royal tombs. The tunnel was advanced from both ends and a sharp Z-bend near mid-length probably marks the point of break-through. A stone memorial tablet found in 1880 provides, in one of the oldest example of cursive Hebrew script, a brief account of the tunnel and the occasion of break-through, which reads in translation: ‘While the hewers yielded the axe, each man towards his fellow, and while there were still three cubits to be hewn, there was heard a man’s voice calling to his fellow, for there was a crack in the rock on the right and on the left.’

The qanats of the Middle East, an art which survived over the centuries with little change, required a greater degree of understanding of the need for an internal lining to provide support and a watertight invert. Joseph Needham (1971) finds evidence of similar tunnels in China as early as 280 BC and conjectures as to whether the art may have passed from China to the Middle East or vice versa. The qanats demanded a degree of conceptual thinking since, as illustrated in Figure 1.1, water was intercepted from springs beneath the surface of the ground, derived from the face of a range of hills, the tunnels serving as conduits to wells, or to surface discharge as ground levels permitted, to satisfy requirements for irrigation. The qanats were constructed from generally vertical shafts at centres of about 150 feet (45 m), their courses marked by rings of spoil around each shaft. Intermediate access for water could be by way of stepped inclined shafts. The water supply to several cities of Asia Minor used ancient canals, partially tunnelled, of considerable length. Arbil in North Iraq, for example, continues to rely upon such a canal of Sennacherib, 20 km long with a tunnel whose ashlar entrance portal is 2.7 m wide, with floor and walls faced in stone slabs.

Figure 1.1 Qanat.

The Greeks and Romans were familiar with the use of tunnelling to undermine (‘sap’) the walls of defended cities, and in the appropriate countermeasures. Many tunnels were also constructed as part of the aqueducts supplying water to cities, such as that for Athens, remaining in use to the present day.

The Greek water supply tunnel on the island of Samos (Plichon 1974) highly regarded by Herodotas, merits a brief description. The tunnel, conceived by Eupalinos and constructed in 525 BC, provided water from a spring to the town of Samos. The tunnel has the form of an approximately square heading 1.8 m×1.8 m, the roof generally provided by a stratum of competent rock. In order to maintain a continuous gradient for the water conduit, formed in tiles, a 0.8 m wide trench was sunk from the heading to depths between 1.5 m and 8 m, a series of stone slabs above the conduit supporting backfill in the trench of spoil from the tunnel.

The Romans have left such fine examples of masonry vaulting that it is no surprise that, for ground capable of standing unsupported over short lengths for a short time, they have demonstrated a capability for building arched tunnels of appreciable span. Their permanent stability depends in the same manner on transmission of load around the arch approximately transverse to the joints between voussoirs. Drainage tunnels mentioned by Livy include the 2210 m (1500 Roman passi) long tunnel to drain Lake Albanus. The tunnel, about 1.5 m wide by 2.5 m high, was constructed in hard volcanic rock in the year 359 BC, using working shafts at intervals of about 45 m. The most notorious Roman drainage project is that of the Lake Fucinus (now Celano) Emissarium, about 5500 m long, 2.5 m wide and 5 m high, which was driven from working shafts at intervals of about 35 m (a Roman actus), vertical and inclined, with ashlar used for lining in unstable ground, built between 41–51 AD. Suetonius describes the first unsuccessful opening ceremony after which the tunnel had to be regraded, the deferred ceremony leading to flooding of the celebratory banquet alongside the tail-race canal.

Road tunnels through spurs of the Apennines around the Bay of Naples were constructed at Cumae and at Puteoli, by Octavian (later Augustus) respectively about 1000 m and 700 m long, 3 m wide and 3.2 m high. Several tunnels of the ancient world are mentioned by Livy, Pliny the elder, Herodotus, Suetonius and Vitruvius, with many of the references given by Humphrey et al. (1998).

1.3 Development of rationale

The breakdown in urban society with the collapse of the Roman empire caused the general absence of demand for tunnelling in Europe for several hundred years, although small-scale mining continued, using the traditional methods described by Agricola (Georg Bauer) in 1556. Széchy (1970) mentions a 5.6 km long drainage tunnel of the Biber mine in Hungary, started around 1400 as an outstanding example of the time. Techniques of timbered support and the methods of excavation were largely derived from mining practice. The construction of the Col di Tenda tunnel on the road between Nice and Genoa was started by Anne of Lusignan in 1450 but later abandoned (Harding 1981) then restarted in 1782 and once again abandoned in 1794 at a length of 2500 m.

Leonardo da Vinci made numerous proposals, which came to nothing, for navigation canals, including a connection between the rivers Garonne and Aude, subsequently constructed as the Languedoc canal by Pierre-Paul Riquet (Sandström 1963), including the Malpas tunnel (157 m×6.7 m high×8.2 m wide) completed in 1681, using gunpowder to blast the rock, but only lined some years later. The celebrated French engineer, Vauban, was also associated with this project.

Sandström contrasts the subsequent practices in canal building in Britain and in Europe, the former by privately financed canal companies at least cost, the latter, financed by the state, having more substantial dimensions. The first English canal tunnel was that of James Brindley on the Bridgewater canal directly entering Worsley coal mine, opened in 1761 and subsequently extended. This marked the beginning of the canal age, with numerous tunnels on the extending canal network. Brindley’s next achievement was the Harecastle tunnel, 2630 m long, 4.3 m high and 2.7 m wide, on the summit of the Grand Trunk (later the Trent and Mersey) Canal, damaged by mining subsidence and replaced by a parallel tunnel constructed by Telford. The canal tunnels constructed by Outram, Jessop, Rennie, Telford and others set traditional standards for tunnel construction in rock and in weaker ground, using the English method described below.

By the seventeenth century, designers of the earliest canal tunnels in France were beginning to apply principles of graphical statics to the criteria for constructing stable arches. The design of timber support was also being developed from the experience in mines. Salt mines at Wielicka near Cracow in Poland, for example, contain massive timbering from the seventeenth century, protected by the saline atmosphere and providing evidence of a well-developed art.

Temporary timber support for tunnelling evolved in different parts of Europe, principally for canal and the early railway tunnels, attracting regional designations and having characteristics adapted to the particular conditions of the locality. Some of the better known systems, illustrated for example by Sandström (1963) and Harding (1981), had particular characteristics.

The German system provided a series of box headings within which the successive sections of the side walls of the tunnel were built from the footings upwards, thus a forerunner of the system of multiple drifts. The method depends on the central dumpling being able to resist without excessive movement pressure transmitted from the side walls, in providing support to the top ‘key’ heading prior to completion of the arch and to ensuring stability while the invert arch is extended in sections.

The Belgian system started from the construction of a top heading, propped approximately to the level of the springing of the arch for a horseshoe tunnel. This heading was then extended to each side to permit construction of the upper part of the arch, which was extended by underpinning, working from side headings. The system was only practicable where rock loads were not heavy.

The English system also started from a central top heading which allowed two timber crown bars to be hoisted into place, the rear ends supported on a completed length of lining, the forward ends propped within the central heading. Development of the heading then allowed additional bars to be erected around the perimeter of the face with boards between each pair to exclude the ground. The system is economical in timber, permits construction of the arch of the tunnel in full-face excavation, and is tolerant of a wide variety of ground conditions, but depends on relatively low ground pressures.

The Austrian system required a strongly constructed central bottom heading upon which a crown heading was constructed. The timbering for full-face excavation was then heavily braced against the central headings, with longitudinal poling boards built on timber bars carried on each frame of timbering. As the lining advanced, so was the timbering propped against each length to maintain stability. The method was capable of withstanding high ground pressures but was particularly extravagant in the demand for timber.

In the absence of other than primitive means for foreseeing the nature of the ground ahead of the advancing tunnel, there were frequent surprises. Linings were in masonry or brickwork depending on local availability of supply. Some of the early railroad tunnels in North America were lined in timber, with long, shaped voussoirs supported on vertical side members.

The first sizeable tunnel in soft ground is recorded by Sandström (1963) as the Tronquoy tunnel on the St Quentin canal in France in 1803, where the method of construction, based on the use of successive headings to construct sections of the arch starting from the footing, was a forerunner to the German system described above.

1.4 New methods, tools and techniques

Rock tunnelling was revolutionised by the first use of explosives for mining in Germany in the seventeenth century (Sandström 1963) and by the use of compressed air for pneumatic drills in 1861. Previously, the methods had been laborious and slow, scarcely improving on principles used by the Egyptians a...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Preface
• Notation
• Table of Contents
• Introduction
• Chapter 1: Background to modern tunnelling
• Chapter 2: Design: the ubiquitous element
• Chapter 3: Planning
• Chapter 4: Studies and Investigations
• Chapter 5: Design of the tunnel project
• Chapter 6: Design of construction
• Chapter 7: Management
• Chapter 8: Hazards, Disputes and their Resolution
• Chapter 9: Coda: the Heathrow Tunnel collapse
• References